Research ArticleAchieving of High Density/Utilization of Active Groups viaSynergic Integration of C=N and C=O Bonds for Ultra-Stableand High-Rate Lithium-Ion Batteries

Tao Sun,1,2 Zong-Jun Li,3 and Xin-Bo Zhang1,⋆

1State Key Laboratory of Rare Earth Resource Utilization, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences,Changchun, 130022, China2University of Chinese Academy of Sciences, Beijing, 100049, China3State Key Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences,Changchun, 130022, China

⋆Correspondence should be addressed to Xin-Bo Zhang; xbzhang@ciac.ac.cn

Received 31 August 2018; Accepted 21 October 2018; Published 16 December 2018

Copyright © 2018 Tao Sun et al. Exclusive Licensee Science andTechnologyReviewPublishingHouse. Distributed under aCreativeCommons Attribution License (CC BY 4.0).

Organic electrode materials are receiving ever-increasing research interest due to their combined advantages, including resourcerenewability, low cost, and environmental friendliness. However, their practical applications are still terribly plagued by lowconductivity, poor rate capability, solubility in electrolyte, and low density/utilization of active groups. In response, herein, asa proof-of-concept experiment, C=N and C=O bonds are synergically integrated into the backbone of pentacene to finely tunethe electronic structures of pentacene. Unexpectedly, the firstly obtained unique 5,7,11,14-tetraaza-6,13-pentacenequinone/reducedgraphene oxide (TAPQ/RGO) composite exhibits superior electrochemical performances, including an ultra-stable cycling stability(up to 2400 cycles) and good rate capability (174 mAh g−1 even at a high current density of 3.2 A g−1), which might be attributedto the abundant active groups, 𝜋-conjugated molecular structure, leaf-like morphology, and the interaction between TAPQ andgraphene.

Lithium-ion batteries (LIBs) are generally proposed to meetthe ever-increasing requirements of electrochemical energystorage for portable electronics and electrical vehicles; how-ever, there have been great concerns on the costs correlatedwith mineral exploitation for LIBsmanufacturing, which stillheavily rely on unsustainable and unbiodegradable inorganiccompounds [1–5]. In sharp contrast to inorganic electrodematerials, organic compounds, characterized by abundantresources, structural diversity, and designability provide analternative for developing sustainable energy materials anddevices [6–11]. However, in the ocean of organic compounds,how to discover the potential high performance organicelectrode material is still a daunting challenge [12–18]. The-oretically, taking full advantage of molecular design anddeepen the understanding of the relationship between struc-ture and property will speed up the research process on theexploration of advanced organic electrodematerial. However,up to now, how to realize the controllable fabrication frommolecular design to excellent battery performances and even

obtain practical application in full cell are still of paramountchallenge.

As an important family of organic molecule, pentacenehas been considered as a benchmark molecular in organicfield effect transistors [19–22].Theoretically, the introductionof different functional group into the planar skeleton ofpentacene will endow its derivatives with unexpected prop-erties and thus functionalized pentacene might be employedas important building blocks for organic electrode mate-rials [23–25]. Especially, when the electroactive sites areintroduced into the 𝜋-conjugated pentacene backbone, theresultant N-heteropentacenequinones will be the potentialelectrode material. However, in fact, to obtain a satisfactoryelectrochemical performance, several challenges need tobe conquered—what kind of functional group should bechosen and how many active sites can be integrated. Moreseriously, like other organic molecular materials, the intrinsiclow conductivity and dissolution problem in electrolyte willhinder its cycling stability and rate capability [26–30]. Up to

AAASResearchVolume 2018, Article ID 1936735, 10 pageshttps://doi.org/10.1155/2018/1936735

2 Research

now, there is no report on N-heteropentacenequinones as anexemplification to display the molecular design of organicelectrode material, to say nothing with good cycling stabilityand rate capability.

Herein, as a proof-of-concept experiment, to system-atically and finely tune the electronic structure of pen-tacene derivatives, C=N and C=O bonds are synergicallyinserted into the backbone of pentacene to achieve acooperative effect and exert the optimum performances,which is then further integrated with reduced grapheneoxide (RGO) to provide an additional energy barrieragainst the intrinsic dissolution of the pentacene-derivativesin electrolyte. Unexpectedly, the resultant novel 5,7,11,14-tetraaza-6,13-pentacenequinone/RGO (TAPQ/RGO) com-posite shows superior electrochemical performances, includ-ing an ultrastable cyclability (193 mAh g−1 after 2400 cyclesat current density of 500 mA g−1) and rate capability (174mAh g−1 at current density of 3200 mA g−1), which is rarelyreported for organic small molecular electrode materials.More importantly, the stable full cell holds superior elec-trochemical performance (92% capacity retention after 220cycles), possessing potential appliance value of TAPQ/RGO.

To develop high performance electrode material, thedesign of organic molecular should take full considerationof high redox potential and specific capacity. In the pursuitof high working potential, the introduction of electron-withdrawing group can efficiently lower the LUMO energies.On the other hand, bridging the carbonyl units with alkylchains is a common method to obtain high specific capacity[7]. However, the inevitable introduction of inactive groupswill increase the molecular weight and reduce the overallenergy densities. In addition, each strategy can only meet oneaspect of the requirements of high-performance electrodematerial. Therefore, a functional integration design strategyshould avoid these disadvantages and try to meet the require-ments for high-performance electrode material as much aspossible, thus to achieve the high utilization of functionalgroups.

As a platform molecular, pentacene with an extended 𝜋-conjugated system can efficiently disperse negative chargeby delocalization. Therefore, the systematic incorporation ofC=N and C=O bonds into the backbone of pentacene mightprovide an exemplification to realize the above design idea.Besides being the active center to bound with lithium ion,the functions of C=N bond are far from being explored [14–17]. Under the premise of ensuring conjugation, replacingthe aromatic carbon with sp2 hybridization nitrogen atomwill increase the redox potential with minimal change of themolecular weight (a tiny increase of 2 g mol−1). In contrast,the incorporation of C=O bond will result in extra increaseof molecular weight of 16 g mol−1. Therefore, it is moreeconomical to introducemore C=Nbond into the conjugatedskeleton to form heteroaromatic system. However, it does notmean the unimportance of C=O bond. On the one hand, toachieve a high capacity, introducing more functional groupsinto conjugation backbone can offer amultielectron reaction.On the other hand, the collaborative effect of C=O andC=N bond plays a unique role in the stabilization of anion,

while still undeveloped in the design of electrode materials.Actually, a rational devise of different active center in thearomatic system can exert the unexpected effectwhich cannotbe solely achieved by the single active functional group.

To examine the effect of different functional groups onits electrochemical performances, three kinds of pentacenederivatives were investigated as cathode material throughthe systematic insertion of C=N and C=O bonds into thebackbone of pentacene (Figure 1(a)). The correctness ofstructure has been verified by various characterizations, andconsistent with previous report thus could be the evidencefor the reliable molecules [23]. Relevant characterizations canbe found in Figure S1-S10 and Table S1, which include theFTIR, 1H-NMR, 13C-NMR, MS, elemental analysis, UV-Vis,and XPS. In front of electrochemical test, a prospective studyof the structure characteristics of derivatives was performedthrough the density functional theory (DFT) calculations.For organic semiconductor, the lowest unoccupied molecularorbital (LUMO) energy level is closely related to its redoxpotential [31, 32]. Compared with other three compounds,TAPQ owns two C=O bonds and four C=N bonds. Thesynergic integration of these electron-withdrawing groups isexpected to induce TAPQ with lower LUMO levels (-3.41 eV)and exhibits higher discharge potentials, which cannot besolely achieved byC=NorC=Obond. Figure 1(b) displays thehighest occupied molecular orbital (HOMO) plots of TAPQand TAPQ∙−; it can be found that the electron in neutralTAPQmainly localized around the conjugated system whichcomposed of C=N and C=O bond. However, the delocaliza-tion of electron in monoanion TAPQ∙− is more uniform andwider on the surface of molecular plane. With the help ofconjugated system and the synergic integration of differentgroups, the reduced species (TAPQ∙−) realize stabilizationand can thus present a theoretical interpretation for thestructural superiority of TAPQ. This effect is more obviousin the HOMO plots of TAPQ-4Li, wherein the HOMO plotsstill retain within the molecular backbone, indicating that theconjugation structure is able to support so much negativecharge (Figure 1(b)). When the neutral TAPQ is reduced tomonoanion TAPQ∙−, the natural bond orbital (NBO) chargeof nitrogen and oxygen atoms drastically increased and thuswill be an active center to bind with the electron-deficientLi cation (Figure S12, Supporting Information) [33]. Due tothe well-designed molecular structure, oxygen and nitrogenatoms in the TAPQ are suitably located for coordination to Liions, which can efficiently stabilize the negative charge. Suchunique chelating coordination environment is impossible inother single active center pentacene derivatives.

To verify the theoretical predictions, the electrochemicalbehavior of three pentacene derivatives was investigated.Obviously, pristine pentacene cannot deliver any capacity forthe lack of any active functional group (Figure 2(a)). In sharpcontrast, when carbonyl (C=O) is introduced into pentacene,the resulting 6,13-pentacenequinone (PTQ) exhibits muchimproved battery performance.The charge/discharge profilesof PTQ contain evident voltage plateaus with average poten-tials of 2.32 and 1.75 V, respectively (Figure 2(b)). However,the initial discharge capacity of PTQ is only 44 mAh g−1 and

Research 3

C=O C=O

PTQ

C-NHC=N C=N

DHTAP

TAPQ

PEN

O

O

O

O

NH

H

N

NN

N

N

N

N

−2

−3

−4

−5

−6

−7

−8

Ener

gy le

vel (

eV)

LUMO

HOMO

TAPQ-4Li

neutral TAPQPEN PTQ DHTAP TAPQ

monoanion TAPQ∙-

+ e-

-. kcal mol-1red,

coord,

-73.48 kcal mol-1 + Li+

-0.8890.882-0.767

(a) (b)

(c)

Figure 1: Density functional theory (DFT) calculations and redox mechanism. (a) Illustrated structures of pentacene derivatives studied inthis work. (b) Energy level diagram of pentacene derivatives and the HOMO plots of TAPQ, TAPQ∙−, and TAPQ-4Li. (c) Redox process ofTAPQ-Li.

decays rapidly in the subsequent cycling process and decreaseto 33 mAh g−1 after 10 cycles.

As a nitrogen-rich heteroacenes, 5,14-dihydro-5,7,12,14-tetraazapentacene (DHTAP) owns a pair of C=N bond inthe pentacene skeleton which determines its potential elec-trochemical performance. The discharge curve of DHTAP issloping in the potential range of 3.0–2.3 V, but the chargeprocess is relatively complicated, indicating the multisteptransformation during the delithiation process (Figure 2(c)).Even though four nitrogen atoms are inserted, the molecularweight of DHTAP (286.12 g mol−1) is still lower than that ofPTQ (308.08 g mol−1). Except a higher theoretical specificcapacity, the practical capacity of DHTAP is also superiorto PTQ, which embody the advantages of heteroaromatic

structure. Although the initial discharge capacity reaches 133mAh g−1, it reduces rapidly at the second cycle and decreasescontinuously in the succeeding cycles.

It is apparent that the pentacene with C=N bond deliversa higher average discharge/charge voltage and more capacitythan the pentacene with C=O bond does. But there is stillroom for further improvement of electrochemical perfor-mances. To a certain extent, DHTAP can be viewed as aphenazine ring connected to a benzene ring by two NH;however, these two sp3 hybridization nitrogen atoms breakthe delocalization of the 𝜋-system in the pentacene skeleton(Figure S13, Supporting Information). For this purpose,a compacted multielectron reaction conjugated structure(TAPQ) is developed. On the one hand, TAPQ owns more

4 Research

4.0

3.5

3.0

2.5

2.0

1.5

0 2 4 6 8Specific capacity (mAh g )

1 st2 nd10 th

Volta

ge (V

vs L

i/L

i)+

-10 10 20 30 40

4.0

3.5

3.0

2.5

2.0

1.5

Volta

ge (V

vs L

i+/L

i)

1 st2 nd10 th

Specific capacity (mAh g )-1

0 40 80 120

4.0

3.5

3.0

2.5

2.0

1.5

Volta

ge (V

vs L

i+/L

i)

1 st2 nd3 th

Specific capacity (mAh g )-10 50 100 150 200 250

4.0

3.5

3.0

2.5

2.0

1.5

Volta

ge (V

vs L

i+/L

i)

1 st2 nd3 th

Specific capacity (mAh g )-1

0

dQ/d

V

1.5 2.0 4.02.5 3.0 3.5Voltage (V)

−2.5 −2.0

−0.6

−0.3

0.0

0.3

−1.5 −1.0 −0.5

Curr

ent (

mA

)

Voltage (V vs Fc+/Fc)

Curr

ent (

mA

) 0.01

0.00

0.01

0.021.5 2.0 2.5 3.0 3.5

Voltage (V vs Li+/Li)

1 st 2 nd4 th3 rd

(a) (b)

(c) (d)

(e) (f)

(g)

Figure 2: Electrochemical performance of pentacene derivatives. Discharge/charge profiles of (a) PEN, (b) PTQ, (c) DHTAP, and (d) TAPQin the voltage range of 1.3–4.0 V at a current density of 50 mA g−1. (e) Differential capacity (dQ/dV) plots of TAPQ. (f) CV curves of TAPQmeasured in DMF solutions containing TBAP as electrolytes. (g) CV curves of TAPQ in coin battery.

active sites will provide more capacity. On the other hand, auniform delocalization of 𝜋 electron on the surface of con-jugated molecular skeleton is beneficial to the stabilizationof the reduced TAPQ species (Figure 1(b)). Based on theabove assumption, the electrochemical behavior of TAPQ

was investigated. As the discharge/charge profiles shownin Figure 2(d), the electrochemical behavior of TAPQ isentirely different from that of DHTAP and PTQ.The evidentdischarge plateau at 2.8 V, 2.1 V and 1.6 V (Figure 2(d)) is wellconsistent with the reduction peaks in cyclic voltammogram

Research 5

N19-O1-N2most stable

N18-O17-N2-O1 N18-O17-N19-O1+9.66 kcal mol +. kcal mol-1 -1

(a)

(b)

1 = -. kcal mol-12 = -36.52 kcal mol-1

3 = -77.81 kcal mol-14 = -78.88 kcal mol-1

+e-,+，Ｃ+ +e-,+，Ｃ+

+e-,+，Ｃ+ +e-,+，Ｃ+

N N N N

NNN

N

NN

N

N

N N

N N

O

O O

O

O

O

O

O

Li

Li

Li

Li

LiLi∙

∙

N N

N N

O

O

Li

Li Li

Li

Figure 3: DFT calculation on TAPQ during redox reaction. (a) Three possible configurations of TAPQ-2Li. (b) The stabilization energies(ΔE) at various stages of the lithiated TAPQ.

(CV) curves (Figure 2(g)). The overall redox behaviors ofdifferential capacity (dQ/dV) and CV curves are similar andcomposed of two distinct regions (Figures 2(e) and 2(g)).

To probe the reaction process, the mechanism is inves-tigated from the aspects of reaction kinetics and thermody-namics. As a matter of fact, lithium ion is bound to both thecarbonyl oxygen and imine nitrogen during the reductionprocess.The calculations indicate that a five-membered hete-rocyclic (C-N-Li-O-C) will be formed during the reductionof TAPQ. NBO charge of TAPQ-Li revealed that even thenegative charge on the nitrogen (-0.767) is inferior to oxygen(-0.889) and the interaction is basically equivalent when theybound with lithium ion which possesses a great positivecharge (0.882) (Figure 1(c)).These conclusions agree with theanalysis of bond length. Except the similar distance fromlithium (Li33-O17 1.745 A; Li33-N3 1.886 A), the increasedbond length of C=O and C=N (Figure 1(c) and Table S2,Supporting Information) implies the simultaneous reductionof active bond. Electron density difference map (EDDM)is a useful technique for the characterization of chemicalbond. For the synergetic coordination, the evident electronictransfer from oxygen and nitrogen to the neighbor lithiumions is compelling evidence. In Figure 1(c), the electronicdensity exhibits an accumulation close to lithium ions anddepletion around imine nitrogen and carbonyl oxygen, whichindicates the obvious interaction between them.

From the viewpoint of stabilization, the monoanion-TAPQ∙− can be significantly stabilized when the lithium

ion is placed in the chelating coordination environ-ment. As shown in Figure 1(c), the coordination energies(Δ𝐸coord,1 -73.48 kcalmol−1) aremore remarkable than reduc-tion energies (Δ𝐸red,1 -48.82 kcal mol−1), implying theimportance of collaborative coordination effect. Through thesynergic integration of different active sites, the simultaneouscoordination of N and O to Li (N-Li-O) is more effective thanthe similar chelating effect of O-Li-O and N-Li-N. Due to thestructural symmetry, three possible configurations can beobtained when the second lithium ion inserted (Figure 3(a),Figure S14-S15, Supporting Information). Placing the secondlithium ion near the fresh carbonyl oxygen will result inan increased energy. Due to the advantage in energy, twolithium ions binding with the same oxygen (N19-O1-N2) arethe most stable configuration.

Thanks to the collaborative coordination center, it ishypothesized that four lithium ions can be stored in TAPQ.This conclusion is supported by thermodynamics calcu-lations wherein the negative stabilization energies valueindicates a favorable binding of the fourth lithium ion(Figure 3(b), Figure S16 and Table S3, Supporting Informa-tion). Nevertheless, the stepwise binding energies are quitedifferent. Revealed in the CV curves, the first two-reductionwave is overlapped, the third and fourth reduction peak areseparated (Figure 2(g)). In fact, careful observation of CVcurves found that the first reduction wave consists of twowaves and exhibits a strong current density. The reversiblefour-electron redox process is demonstrated when the CV

6 Research

400

300

200

100

01600 1400 1200 1000 800 4.0 3.5 3.0 2.5 2.0 1.5

Voltage (V)Wavelength (cm-)

Abs

orba

nce (

a.u.

)

C=OC=N

)A

ccum

ulat

ed ca

paci

ty (m

Ah

g-1

536 534 532 530 528 526Binding energy (eV)

Inte

nsity

(a.u

.)

Recharged

Discharged

As-prepared

Binding energy (eV)

Inte

nsity

(a.u

.)

Recharged

Discharged

As-prepared

N-LiC=N

406 404 402 400 398 396 394

Binding energy (eV)

Recharged

294 291 288 285 282 279

C-NC=NC=O

C=C,C-C,C-H

Inte

nsity

(a.u

.) Discharged

Binding energy (eV)294 291 288 285 282 279

C-Li

C-NC=N

C=O

C=C,C-C,C-H

Inte

nsity

(a.u

.) As-prepared

Binding energy (eV)294297 291 288 285 282 279

C-NC=NC=O

C=C,C-C,C-H

Inte

nsity

(a.u

.)

(a) (b) (c)

(d) (e) (f)

Figure 4: Ex-situ analyses of TAPQ electrode at different states. (a) Ex-situ FTIR characterizations. (b) Ex-situ XPS local scan spectra of O1s regions. (c) Ex-situ XPS local scan spectra of N 1s regions. (d-f) Ex-situ XPS local scan spectra of C 1s regions.

curves of TAPQwere measured in DMF solutions containingTBAP as electrolytes (Figure 2(f)).

In combination with DFT calculations, ex situ Fouriertransform infrared spectroscopy (FTIR) and X-ray photo-electron spectroscopy (XPS) were employed to identify thestructural changes. When discharged to 1.3 V, the stretchingvibration of C=O bond located at 1696 cm−1 disappeared(Figure 4(a)). The vanished C=O bond restored in the nextcharge process. The stretching vibrations of C=N bonds areobserved at 1534, 1512, 1482, 1463 cm−1. During the lithiationprocess, the intensity of C=N bond gradually weakens. In thecharge process, these peaks reversibly appear, which suggeststhe participation of C=N groups. The repeatability FTIRspectral signals in other regions also provide compellingevidence for the reversible electrochemical behaviour ofTAPQ. Similarly, the reversible disappearance and formationof the characteristic C=O and C=N bond are also detected inthe ex-XPS. In the discharged state, the O 1s band shift towardlower binding energy indicates the participation of C=O inthe reaction (Figure 4(b)). Due to the formation of a newbond between N and Li atoms (398.2 eV), the characteristicpeak of C=N bond become wider in the discharged state(Figure 4(c)). The simultaneous coordination of oxygen andnitrogen atom to lithium is also confirmed by the C1s spec-trum (Figures 4(d)–4(f)). Accompanied by the weakeningof C=O and C=N bond, a new peak (290.35 eV) indexed toC–Li appears during discharge process, which results fromthe lithiation reaction with active groups. In the rechargedelectrode, C-Li peak disappears and exhibits the reversibletransformation of C=O and C=N groups.

More active sites mean more capacity, and TAPQ indeedprovides more capacity as expected. The specific dischargeand charge capacity of TAPQ are 237 and 232 mAh g−1during the first cycle, respectively (Figure 2(d)). Imperfectly,there is a serious capacity fading in TAPQ when it acts asorganic cathode (Figure S17-S18, Supporting Information).As is known, the capacity fading is mainly caused by tworeasons [7, 9]. One is the intrinsic electrochemical instability,and the other is the dissolution of active material into theorganic electrolyte [34, 35]. To explore the reason of capac-ity fading, relevant experiments were carried out. Firstly,the disassembled separator becomes green when TAPQ isreduced to anion species which indicates that the dissolu-tion occurred (Figure S17, Supporting Information). Eventhough the capacity decays seriously, the discharge/chargeprofiles maintain well and voltage plateaus are still evident,which indicate the electrochemical behavior of TAPQ is notchanged and the electrochemical reaction is still reversible(Figure S18, Supporting Information). Additionally, the typi-cal FTIR peaks of TAPQ in the cycled electrode are consistentwith the pristine samples which further testify the intrinsicchemical stability of TAPQ (Figure S17, Supporting Infor-mation). Therefore, the capacity fading of TAPQ electrodeshould be attributed to the dissolution rather than its intrinsicelectrochemical instability.

In response, coupling TAPQ with graphene will be thepossible solution to tackle the dissolution and improve theconductivity. On the one hand, the charge transport of TAPQis along the 𝜋–𝜋 stacking direction of the aromatic rings.As an n-type organic semiconductor, the electron mobility

Research 7

(a) (b)

(c) (d)

Figure 5: Morphology of TAPQ and its composite. (a) SEM images of bulk TAPQ. (b) SEM images of TAPQ nanoleaf. (c) SEM images ofTAPQ/RGO. (d) TEM images of TAPQ/RGO.

of TAPQ is in the range of 0.04−0.12 cm2 V−1s−1 [36], itsintrinsic conductivity is far less than the requirements ofhigh-performance electrode materials. Therefore, it is nec-essary to introduce conductive carbon and form compositeelectrodes to solve the poor electrical conductivity. On theother hand, the interactions between organic molecule andgraphene are strong physisorption. The introduction of moreactive group into the pentacene will not only achieve ahigh capacity but also enhance the binding energies betweenorganic molecule and graphene due to the incorporation ofheteroatoms with strong electronegativity. In the dissolutionprocess, an additional energy barrier needs to be overcomewhen the organic molecule desorbs from graphene, thereforethe strong physisorption will slow down the dissolution rate.However, previous reported ultrasonic method is hard toachieve a uniform TAPQ/graphene composite (Figure S19,Supporting Information).

To tackle the above problems, in situ composite approachwas provided. When the DMF solution of TAPQ is quicklyinjected into ether, TAPQ molecular will undergo a sec-ondary aggregation process and precipitate from the poorsolvent. The 𝜋-conjugated planar structure makes TAPQ hasa favor to assemble into 2D structure and thus achieves

the transformation from bulk particle to nanoleaf (Figures5(a) and 5(b)). With the size decreasing to micro/nanometerscale, TAPQ nanoleaf will provide more active sites and facileelectronic/ionic transfer and diffusion. Accordingly, whenthe mixture solution of GO and TAPQ was injected intoether, the leaf-like TAPQ will be formed immediately on thesurface of GO. Meanwhile, the smooth GO layer will becomeshrinkage in the ether, thus coating the TAPQ nanoleaf. Inthe composite, TAPQ nanoleaf will absorb tightly on thesurface of RGO layer for their similar 2D structure (Figures5(c)-5(d)). Such plane-to-plane contact mode allows a morefacilitated electron transfer between TAPQ and graphenelayer which is favorable for the stabilization of the reducedTAPQ species. Tethering the electro-active molecules on thesurface of RGO through the secondary precipitation method,not only the morphology of TAPQ is changed from thebulk particle to nanoleaf but also allow a closer contactbetween TAPQ and RGO, thus enhancing the conductivityand lithium-ion accessibility of the electrode, which canget supports from electrochemical impedance spectroscopy(Figure S20, Supporting Information).

To highlight the superiority of in situ composite method,a series of comparison experiments were carried out. From

8 Research

(a) (b)

(c)

Spec

ific c

apac

ity (m

Ah

g)

0 20 40 60 80 100Cycle number

TAPQ/RGO TAPQ/RGO-STAPQ nanoleaf bulk TAPQ

300

200

100

0

-1

Spec

ific c

apac

ity (m

Ah

g)

Cycle number

50300

100

200

400 800 1600 3200200 50

100

0

0 10 20 30 40 50 60 70 80

mA g-1

-1

Spec

ific c

apac

ity (m

Ah

g)

300

250

200

150

100

50

00 400 800 1200 1600 2000

Cycle number

Charge Discharge

TAPQ/RGO

TAPQ/RGO-S

2400

-1

Figure 6: Electrochemical performance of TAPQ and its composites. (a) Cycling performance of TAPQ and its composites at a currentdensity of 50 mA g−1. (b) Rate performance of TAPQ and its composites. (c) Long-term cycling stability of TAPQ/RGO and TAPQ/RGO-Sat a current density of 500 mA g−1.

the start of cycling, the discharge capacity of bulk TAPQdecays rapidly from 238 to 90 mAh g−1 at 20th cycle(Figure 6(a)). This situation can be also found in the rateperformance, wherein the bulk TAPQ completely lost thecapacities when the current density merely increased to 400mA g−1 (Figure 6(b)). Compared with the bulk particle, ateach current density, TAPQnanoleaf all exhibits an enhancedcapacity for its large specific surface areas and more exposedactive sites. However, there is also a serious capacity fading asthat in bulk TAPQ (Figure 6(a)). Fortunately, this problem issolved after the TAPQ is compositedwith RGO. TAPQ/RGO-S, which obtained from the sonication treatment of RGOand TAPQ, delivers stable capacities of 245, 232, 219, 205,184, 168, and 161 mAh g−1, respectively, as the currentdensities increased from 50 to 3200 mA g−1 (Figure 6(b)).Regretfully, the gradually decreased trend of the capacity canbe also found in TAPQ/RGO-S, from the initial dischargeat 242 mAh g−1 decline to 146 mAh g−1 after 100 cycles(Figure 6(a)). In comparison, TAPQ/RGO that prepared

from in situ composite route exhibits better electrochemicalperformance than its sonication treated counterpart. Atthe same charge/discharge current, the reversible capacityof TAPQ/RGO is higher than that of TAPQ/RGO-S. Thedischarge capacity of TAPQ/RGO even retained at 221, 205,and 174 mAh g−1 when cycled at 800, 1600, and 3200 mAg−1 (Figure 6(b)).More importantly, the rapid capacity fadingcannot be found in TAPQ/RGO. When cycled at 50 mA g−1,the reversible capacity reaches 241 mAh g−1 and sustains at212 mAh g−1 at the 100th cycle (Figure 6(a)). As the currentdensity increased to 500 mA g−1, the advantage of in situcomposite method can be evidently revealed (Figure 6(c)).The reversible capacity of TAPQ/RGO remains at 193 mAhg−1 after 2400 cycleswith a capacity retention of 80%,whereasTAPQ/RGO-S only achieves at 60% after 1770 cycles. Such astable cycling performance is rarely seen in organic electrodematerials (Table S4, Supporting Information).

Encouraged by the superior electrochemical perfor-mances presented above and to reveal the potential practical

Research 9

value of TAPQ/RGO, a Li-ion full cell with TAPQ/RGO asthe cathode and graphite as the anode was assembled. Thedischarge/charge behavior of full cell remains the same asthat of half-cell, and delivers a reversible capacity of 243mAh g−1 at 50 mA g−1 with an average operation voltage of2.5 V (Figure S23a, Supporting Information). The prominentfeature of this full cell is the stable cycling performance.Cycled at a constant current of 100 mA g−1, the full celldelivers a stable capacity around at 225 mAh g−1 and thebattery remained 92% of its initial capacity after 220 cycles(Figure S23c, Supporting Information).

In summary, to achieve the high energy density and uti-lization of functional groups, a series of pentacene derivativesare devised through a synergic integration of C=N and C=Obonds into the conjugated skeleton. Based on substantialDFT calculations and characterization techniques, the redoxmechanism of TAPQ is investigated from different perspec-tives. To overcome the high solubility and low conductivity,TAPQ/RGO composite is synthesized through a convenientin situ mixture method and exhibits an ultra-stable cyclingstability and a good rate capability. In this work, the conceptof functional integration design strategy presents a paradigmon the design of new materials, which not only can achievethe quick discovery of potential high performance organicelectrode material but also provide a new strategy on theimprovement of battery performance based on the takingfull advantage of molecular design. We anticipate that thesepreliminary investigations will encourage researchers to paymore attention on the structure-performance relationship,thus to have a deeper understanding on the design of organicelectrodes.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was financially supported by the National Natu-ral Science Foundation of China (21725103 and 51472232),JCKY2016130B010, Jilin Province Science and Technol-ogy Development Plan Funding Project (20180101203JC,20160101289JC), and Changchun Science and TechnologyDevelopment Plan Funding Project (18DY012).

Supplementary Materials

Scheme S1. Synthesis routes of TAPQ. Scheme S2. Synthesisroutes of PTQ. Figure S1. (a) FTIR spectral of PTQ. (b) XRDpattern of PTQ. Figure S2. 1H NMR (DMSO-d6) spectrumof DHTAP. Figure S3. FTIR spectra of DHTAP. FigureS4. ESI-FTMS of DHTAP. Figure S5. 1H NMR (DMSO-d6) spectrum of TAPQ. Figure S6. 13C NMR (DMSO-d6)spectrum of TAPQ. Figure S7. ESI-FTMS of TAPQ. TableS1. Elemental analysis results of pentacene derivatives andTAPQ/RGO. Figure S8. UV-vis spectra of TAPQ. FigureS9. XPS spectra of the C1s, O1s and N1s peak fittings ofTAPQ. Figure S10. FTIR of bulk TAPQ, TAPQ nanoleafand TAPQ/RGO. Figure S11. TGA of TAPQ and composite.

Figure S12. The natural bond orbital (NBO) charge dis-tributions of (a) the neutral TAPQ and (b) monoanionTAPQ∙−. Figure S13. (a) The highest occupied molecularorbital (HOMO) plot of DHTAP. (b) The lowest unoccupiedmolecular orbital (LUMO) plot of DHTAP. Figure S14. Thepossible configurations of TAPQ-2Li: (a, d) the optimizedstructures; (b, e) HOMOplots; (c, f) LUMOplots of N18-O17-N19-O1 and N18-O17-N2-O1. Figure S15. The most stableconfigurations of TAPQ-2Li: (a) the optimized structures; (b)HOMO plots; (c) LUMO plots of N19-O1-N2. Figure S16.(a, d) the optimized structures; (b, e) HOMO plots; (c, f)LUMO plots of TAPQ-3Li and TAPQ-4Li. Table S2. Themajor bond length change in TAPQ intermediate. The unit ofbond length is A. Table S3. Summary of energies of the redoxintermediate and reaction step. Figure S17. (a) The FTIRspectrum of cycled TAPQ electrode and pristine bulk TAPQ.The insets are the optical photograph of cycled separator. Theleft one corresponds to the bulk TAPQ electrode. The rightone corresponds to the TAPQ/RGO electrode. (b) The solu-bility of TAPQ (left), TAPQ/RGO (right) in the electrolyte.Figure S18. Discharge/charge profiles of TAPQ nanoleaf atcurrent density of 100 mA g−1. Figure S19. SEM imageof TAPQ/RGO-S. Figure S20. Electrochemical impedancespectroscopy of TAPQ and its composites: (a) TAPQnanoleafand bulk TAPQ; (b) TAPQ/RGO and TAPQ/RGO-S after theelectrode discharged to 1.3 V at 5th cycle; (c) TAPQ/RGO at50

th and 100th cycle; (d) TAPQ/RGO-S at 50th and 100th cycle.Figure S21: Electrochemical performances of TAPQ/RGOhalf-cell. (a) CV curves of the initial 3 cycles at 0.1 mV s−1;(b) CV curves of TAPQ/RGO at different scan rates; (c) log(i)versus log(v) plots at different redox states. Figure S22.Charge/discharge curves of TAPQ/RGO (a) and RGO (b) ata current density of 50 mA g−1 in the voltage range of 1.3-4.0V. Figure S23. (a) Charge/discharge curves of TAPQ/RGOfull cell in the voltage range of 1.29 and 3.65 V at 50 mAg−1; (b) Charge/discharge curves of graphite cycling between0.01 and 1.4 V at 50 mA g−1; (c) Cycling performance of LIBsfull cell by using a TAPQ/RGO cathode and graphite anodeat 100 mA g−1. Figure S24. SEM images of TAPQ/RGOelectrodes: a) after 50 cycles, b) after 100 cycles. Comparingthe electrode after cycling, the nanoleaf morphology ofTAPQ is still maintained and well composited with RGOand thus demonstrating the superiority of in situ compositemethod. Figure S25. Voltage-Capacity curves of TAPQ andits composites corresponding to Figures 6(a) and 6(b) at thecurrent density of 50 mA g−1. Table S4. Comparison ofTAPQ/RGO with those organic/carbon composite electrodematerials. (Supplementary Materials)

References

[1] M. Armand and J. M. Tarascon, “Building better batteries,”Nature, vol. 451, no. 7179, pp. 652–657, 2008.

[2] D. Larcher and J. M. Tarascon, “Towards greener and moresustainable batteries for electrical energy storage,” NatureChemistry, vol. 7, p. 19, 2014.

[3] B. Kong, J. Tang, Y. Zhang et al., “Incorporation of well-dispersed sub-5-nm graphitic pencil nanodots into orderedmesoporous frameworks,”Nature Chemistry, vol. 8, p. 171, 2015.

10 Research

[4] R. Ma, S. Kwon, Q. Zheng et al., “Carbon-nanotube/silver net-works in nitrile butadiene rubber for highly conductive flexibleadhesives,” Advanced Materials, vol. 24, no. 25, pp. 3344–3349,2012.

[5] B. Kong, L. Zu, C. Peng et al., “Direct superassemblies offreestanding metal–carbon frameworks featuring reversiblecrystalline-phase transformation for electrochemical sodiumstorage,” Journal of the American Chemical Society, vol. 138, p.16533, 2016.

[6] Z. Song andH. Zhou, “Towards sustainable and versatile energystorage devices: an overview of organic electrode materials,”Environmental Science, vol. 6, no. 12, p. 2280, 2013.

[7] Q. Zhao, Y. Lu, and J. Chen, “Oxocarbon Salts for Fast Recharge-able Batteries,”A Journal of German Chemical Society, 2016.

[8] P. Hu, H. Wang, Y. Yang, J. Yang, J. Lin, and L. Guo, “Renew-able-biomolecule-based full lithium-ion batteries,” AdvancedMaterials, vol. 28, p. 3486, 2016.

[9] Z. Zhu and J. Chen, “Review—advanced carbon-supportedorganic electrode materials for lithium (sodium)-ion batteries,”Journal of the Electrochemical Society, vol. 162, 2015.

[10] J. Wu, L. Zhao, X. Xu et al., “Hydrophobic cysteine poly(di-sulfide)-based redox-hypersensitive nanoparticle platform forcancer theranostics,” Angewandte Chemie; International Edi-tion, vol. 54, no. 32, pp. 9218–9223, 2015.

[11] J. Wu, X. Rui, C. Wang et al., “Nanostructured conjugatedladder polymers for stable and fast lithium storage anodes withhigh-capacity,” Advanced Energy Materials, vol. 5, Article ID1402189, 2015.

[12] J. Xie and Q. Zhang, “Recent progress in rechargeable lithiumbatteries with organic materials as promising electrodes,” Jour-nal of Materials Chemistry A, vol. 4, no. 11, p. 7091, 2016.

[13] J. Xie, P. Gu, and Q. Zhang, “Nanostructured conjugatedpolymers: toward high-performance organic electrodes forrechargeable batteries,” ACS Energy Letters, vol. 2, p. 1985, 2017.

[14] C. M. Elizabeth, C. G. Javier, and A. Michel, “Polymeric schiffbases as low-voltage redox centers for sodium-ion batteries,”Angewandte Chemie; International Edition, vol. 53, p. 5341, 2014.

[15] M. Lopez-Herraiz, E. Castillo-Martinez, J. Carretero-Gonzalez,J. Carrasco, T. Rojo, and M. Armand, “Oligomeric-Schiff basesas negative electrodes for sodium ion batteries: unveiling thenature of their active redox centers,” Energy; EnvironmentalScience, vol. 8, no. 5, pp. 1463–1470, 2015.

[16] M. Lee, J. Hong,D.H. Seo et al., “Redox cofactor frombiologicalenergy transduction as molecularly tunable energy-storagecompound,” Angewandte Chemie, vol. 52, p. 8322, 2013.

[17] J. Hong, M. Lee, B. Lee, D.-H. Seo, C. B. Park, and K. Kang,“Biologically inspired pteridine redox centres for rechargeablebatteries,” Nature Communications, vol. 5, p. 3396, 2014.

[18] I. A. Rodrıguez-Perez, Z. Jian, P. K. Waldenmaier et al., “Ahydrocarbon cathode for dual-ion batteries,” ACS Energy Let-ters, vol. 1, no. 4, pp. 719–723, 2016.

[19] Z. Liang, Q. Tang, J. Liu, J. Li, F. Yan, and Q. Miao, “N-typeorganic semiconductors based on 𝜋-deficient pentacenequi-nones: synthesis, electronic structures, molecular packing, andthin film transistors,” Chemistry of Materials, vol. 22, no. 4, pp.1402–1409, 2010.

[20] C. Wang, J. Zhang, G. Long et al., “Synthesis, structure, andair-stable n-type field-effect transistor behaviors of functional-ized octaazanonacene-8,19-dione,”Angewandte Chemie, vol. 54,p. 6292, 2015.

[21] J. Li and Q. Zhang, “Linearly fused azaacenes: novel approachesand new applications beyond field-effect transistors (FETs),”Applied Materials & Interfaces, vol. 7, no. 30, 2015.

[22] J. Li, J. Miao, G. Long et al., “N-Heteroheptacenequinoneand N-heterononacenequinone: synthesis, physical properties,crystal structures and photoelectrochemical behaviors,” Journalof Materials Chemistry C, vol. 3, 2015.

[23] S. Miao, A. L. Appleton, N. Berger et al., “6,13-Diethy-nyl-5,7,12,14-tetraazapentacene,” Chemistry-A European Jour-nal, vol. 15, 2009.

[24] J. I.Wu, C. S.Wannere, Y.Mo, P. v. R. Schleyer, andU.H. F. Bunz,“4n 𝜋 Electrons but Stable: N,N-Dihydrodiazapentacenes,”TheJournal of Organic Chemistry, vol. 74, no. 7, pp. 2811–2816, 2009.

[25] A. Shimizu, Y. Tsujii, H. Kuramoto et al., “Nitrogen-containingpolycyclic quinones as cathode materials for lithiumion batter-ies with increased voltage,” Energy Technology, vol. 2, no. 2, pp.155–158, 2014.

[26] Y.-X. Yu, “A dispersion-corrected DFT study on adsorption ofbattery active materials anthraquinone and its derivatives onmonolayer graphene and h-BN,” Journal of Materials ChemistryA, vol. 2, no. 5, p. 8910, 2014.

[27] Y.-X. Yu, “Binding energy and work function of organic elec-trode materials phenanthraquinone, pyromellitic dianhydrideand their derivatives adsorbed on graphene,” ACS AppliedMaterials; Interfaces, vol. 6, p. 16267, 2014.

[28] W. Ai, W. Zhou, Z. Du et al., “Toward high energy organiccathodes for li-ion batteries: a case study of vat dye/graphenecomposites,” Advanced Functional Materials, 2016.

[29] H. Wang, P. Hu, J. Yang, G. Gong, L. Guo, and X. Chen,“Renewable-juglone-based high-performance sodium-ion bat-teries,” AdvancedMaterials, vol. 27, p. 2348, 2015.

[30] M. Lee, J. Hong, H. Kim et al., “Organic nanohybrids for fastand sustainable energy storage,” Advanced Materials, vol. 26, p.2558, 2014.

[31] Y. Liang, P. Zhang, and J. Chen, “Function-oriented designof conjugated carbonyl compound electrodes for high energylithium batteries,” Chemical Science, vol. 4, p. 1330, 2013.

[32] Y. Liang, P. Zhang, S. Yang, Z. Tao, and J. Chen, “Fusedheteroaromatic organic compounds for high-power electrodesof rechargeable lithium batteries,” Advanced Energy Materials,vol. 3, p. 600, 2013.

[33] T. Sun, v Li, H.-g. Wang, D. Bao, F.-l. Meng, and X.-b. Zhang,“A biodegradable polydopamine-derived electrode materialfor high-capacity and long-life lithium-ion and sodium-ionbatteries,” Angewandte Chemie, vol. 128, p. 10820, 2016.

[34] Z. Song, Y. Qian,M.Otani, andH. Zhou, “Stable li–organic bat-teries with nafion-based sandwich-type separators,” AdvancedEnergy Materials, vol. 6, Article ID 1501780, 2016.

[35] W. Wang, M. T. Winkler, O. Gunawan et al., “Device charac-teristics of CZTSSe thin-film solar cells with 12.6% efficiency,”Advanced Energy Materials, vol. 4, no. 7, Article ID 1301465,2014.

[36] Q. Tang, Z. Liang, J. Liu, J. Xu, andQ.Miao, “N-heteroquinones:quadruple weak hydrogen bonds and n-channel transistors,”Chemical Communications, vol. 46, p. 2977, 2010.